An implantable cardiac pacer can have various pacing modes as well as various output parameters such as rate, output level, current, voltage, pulse width sensitivity, refractory period, etc. In some cardiac pacers both the mode, e.g, R-wave inhibited VVI, as well as the various output parameters, are preset during production, whereas in other cardiac pacers either mode or output parameters or both can be altered by external control or programming. Such output parameters and/or pacing mode changes are usually accomplished by the attending physician, usually during an office visit. Therefore, such cardiac pacers may not be responsive to the physiological requirements of the patients. Such requirements fluctuate often during a 24-hour period, certainly more frequently than the interval between patient's visits to the physician. Thus the patient must suffer less than optimum heart pacing.
Increased emphasis is being placed on the use of physiological parameters to control the output parameters, and particularly the rate of stimulation, of implantable cardiac pacers. Such physiological parameters can include activity of the patient (Dahl U.S. Pat. No. 4,140,132), sensed ionic changes (Wirtzfeld U.S. Pat. No. 4,202,339), the patient's threshold, i.e., the level of a stimulus pulse required to evoke a resulting heartbeat when the pulse is delivered to the patient's heart (Wittkampf, et al. U.S. Pat. No. 4,305,396), and the stimulus to repolarization of the T-wave interval (Rickards U.S. Pat. No. 4,228,803). The detection of such changes is utilized either to increase or to decrease the rate of stimulation.
Measurement of physical activity or of the ionic level in the blood does not appear to measure the effectiveness of the pulse emitted from the cardiac pacer initiating myocardial contraction.
While the stimulus to repolarization interval on the T-wave can provide information as to stress on the heart, the T-wave is sometimes difficult to detect. This is illustrated in the article by Fananapazir entitled Reliability of the Evoked Response in Determining the Paced Ventricular Rate and Performance of the QT of Rate Responsive (TX) Pacemaker, "Pace", vol. 8, pp. 701-714 (September-October, 1985). In this article, it is stated that problems resulted from attempts to provide T-wave sensing in the long term with patients where the system was being used to control heart pacing rates.
Also, the amplitude, as measured from the base line, of the repolarization potential or T-wave tends to be low, thereby posing problems in detection. In addition, within means for detecting the evoked potential of the T-wave, a time period or "window" for sensing of the T-wave must be established and programmed either at manufacture or after implantation. If there is an improper setting of the sensing window, the T-wave may not be sensed. Also, in certain patients, the T-wave may be bimodal; that is, there may be two peaks rather than one. It is possible that the amplitude of the first peak is not sufficient to be detected, and that the second peak may be outside of the sensing window. Alternatively, neither of the peaks of the bimodal T-wave may be of sufficient amplitude to be detected.
In accordance with this invention, a more reliable technique for detecting stress level changes is provided, particularly in a paced heart where it is desirable to provide feedback means so that the rate of pacing of the heart can vary in a manner responsive to heart stress. Improvements of long term reliability and ease of operation are provided by this invention.